Learning to Learn¶
After writing my post advice for young people, a couple of people asked about my learning process. I could discuss overcoming plateaus or developing mastery, learning for the joy of learning. I could also talk about how to avoid feeling overwhelmed by new topics and break them down into smaller pieces. However, I think that has been done before.
Instead, I'm going to explore a new style. I'm just going to go through a chronological telling of my life and what I learned from just trying new things. I'm going to talk about the tactics and strategies and see how this pans out.
Its really revolves around some core ideas, that I definitely knew before, but didn't see the patterns for myself. Until years into trying stuff, the important of teaching to learn, having process vs outcome goals, planning and periodization, and just not doing anything flashy too early.
I'm going to break this down into two sections:
What I did in high school and college
- Data Science
What I did after college
It's a false dichotomy in the sense that the tactics and strategies for knowledge acquisition apply to physical skills, and the volume and leverage required to learn physical skills also apply to knowledge acquisition. But consider this an experiment, I'll go over each one and a lesson I learned.
Thanks to the following people for reading drafts of this post and providing feedback:
Just wanted to thank a few readers for early feedback and edits:
I grew up in a environment with very few distractions, and it was very easy to learn. My parents weren't around much: when I was a kid (~ 10 years old), both my mom and dad worked two jobs, and no one was at home. The one thing they showed me was that education was the way out of poverty, and that they worked very hard to make sure I had access to it. Every Saturday, my mom would take the time to walk me to the library, where I would borrow 10 books. I was only allowed to use the computer for an hour a day and watch 30 minutes of television per week, basically just one cartoon.
Learning became my escape: I wanted to understand physics, read about igneous and metamorphic rocks, acquire knowledge about different types of dinosaurs and animals.
My parents were not around, but I know the one thing they showed me was that education was the way out of poverty. They worked very hard to ensure I had access to it. This is all to say that I grew up in a place with very few distractions, making it easy for me to learn.
Don't study things you'll be taught¶
These next few sections will mostly be random things I had noticed about my learning through high school and university.
One of the first realizations I made when I was a kid was that if I went to advanced math or science school. That those 4 years would be wasted as I'd only get maybe 1 extra credit in college. It didn't feel like a great exchange rate, so I decide to enroll into an art program instead of a high school stem program.
Just be ahead a little bit¶
I quickly learned that if I spent the first few weeks working two or three times harder than everyone else to learn new software like Adobe Photoshop or Illustrator, I would be far ahead of the class. This boosted my self-esteem and made me feel intelligent. Moreover, it allowed me to gain more practice. Even if neither of us understood what was going on, the fact that someone could ask me a question and I could figure out the answer faster meant that I got to learn and teach simultaneously. This widened the gap between me and my peers significantly, which was basically two reps. Eventually, helping others just meant it gave me access to problems I wouldn't have otherwise encountered.
Something I noticed
I was never the best artist, and I don't think that was the goal for me either. However, I noticed that most of my skilled friends were always drawing and practicing. Even to this day, there are people I know whom I've never seen without a sketchbook: they probably fill 12 to 15 sketchbooks every year. It's always amusing that whenever people admire their work and call them talented, they're unaware that for the past 12 years, every time I've been late for a coffee, every time we've visited a gallery, every time we've gone for a walk, they've always been drawing.
Create your peer group¶
After my second year of college, I decided to learn machine learning and leave physics behind. I had just completed Andrew Ng's machine learning course on Coursera, and the possibilities amazed me. However, I didn't know anyone who shared my interest in machine learning. Just like in high school, I had to push myself hard to get ahead. At Waterloo, I started the Data Science Club and founded a machine learning group called Data Hackers. I even created a guide on how to start your own data science/machine learning club: in just two years, I had a Facebook group of 7000 people who shared my passion for machine learning and data science.
Similar to how I gained a following on Twitter, I would read papers and share summaries. I posted helpful content, attended hackathons, and met other data science club founders to exchange ideas and support each other. I told myself that if I could make Waterloo the top data science school in Canada, my own data science skills would get pretty good.
You only need to be a little bit ahead¶
I found myself in a situation where hundreds of people messaged me with their machine learning and data science questions, not knowing I literally know no better than them. However, I would diligently search Google and YouTube to find answers and call them to discuss the topics. Just like in highschool, although I was only slightly more knowledgeable than them, having this audience allowed me collect interesting questions and find answers.
By the time I started the Data Science Club, my goal was simply to teach as much as possible. I gave a couple of lengthy lectures every semester and organized career panels. I practiced reviewing résumés to develop a better understanding of how to write my own, and even honed my interview skills by conducting mock interviews.
Eventually, I started dedicating half my Saturday to this process. I'd just study with a Google Meet session running in the background: if anyone popped in, I'd learn more about their experiences, answer their questions whenever possible, and learn from their insights. Many of those people are now my friends whom I've known for over 10 years.
Every opportunity to teach, learn, and practice is a chance to improve. Being able to explain something well results in people asking out-of-the-box questions that you wouldn't have thought of. I find that the goal of teaching is secretly for the audience to bring all the idiosyncratic edge cases of your knowledge to your attention so you can learn from them. It's like a knowledge pyramid scheme.
Even if I'm only two weeks ahead, trying to teach means spending a few hours helping them get two weeks ahead. If you do the math, if someone was two weeks ahead of everyone else and spent one hour teaching 30 people, you just provided 60 human weeks of learning which is over a year of value in one hour. That's a pretty good exchange rate.
The real world¶
That basically summarizes how I thought about learning in a school setting where because roughly speaking, everyone is somewhere on the assembly line, the system I used to make progress in new hobbies was very differen since it's rare to enter a new place as an adult and find you have access to a large group of people who are all trying to learn the same thing with the same level of intensity as a group of high achieving students, many of those same people will not pursuit any hobbies as they decide to focus on their careers.
The biggest thing I learned in pottery, except for the fact that I sucked for like two years, was how we measured progress. I just remember thinking that I wanted to go to class and leave with six nice cups. I would make a couple of cups, but they would break somewhere in the process, and that would feel really disheartening. I remember telling my teacher what was going on, and they gave me advice that stuck with me forever. They said, "You should not focus on how many cups you make; you should just count how many bags of clay you go through." This was a huge perspective shift for me because a cup was 250 g, while a bag of clay was 25 kg. If I had considered the bags of clay as the metric, my progress would have been a straight
0-0 for the last 2 months.
Process vs Outcome¶
I was holding the work a little too precious, and I wasn't actually putting in any volume.
What did I learn? I basically learned that we often set outcome-based goals instead of process-based goals, and even when we do set process-based goals, we are definitely not using the right measurement. No wonder I wasn't doing enough to actually achieve the outcomes that I wanted. Moreover, I was also not measuring the right thing, usually the units and order of manitude were off and figuring out the right metric bags of clay was a huge part of the process.
By the time I started weightlifting, I kind of already accepted that I had to focus on process based goals. When I learned about progressive overload, it made sense that if my metric was just weight _ reps _ sets, all I had to do was to increase that number every week. I could increase, weight, reps, or sets, every week, and I would be making progress.
Periodize your training¶
I learned that not only do strength athletes do progressive overload, they also periodize their training. They have a hypertrophy phase, a strength phase, and a peaking phase. Not only that, they take deload weeks! What I got from my time weightlifting is that for anything that I'm trying to learn, not only should I think about volume instead of outcomes, I should also think about how I can do progressive overload to increase volume and periodize the training and apply rest as a feature of the training program. Many professional also both on season and off-season where they can focus on different aspects of their training like skills development vs conditioning.
It's also just more fun to set learning goals and peak over some multi-month period: it makes the expectation of progress more realistic, and it's easier to stay motivated.
I think Jiu Jitsu is by far the most interesting thing I have ever done. There are so many lessons I've learned from practicing it, but I want to focus on the idea of choosing your teachers. There is a huge advantage in choosing to learn from active competitors. I believe that techniques and tips in combat sports, in particular, are very difficult to fake. You can't be taught a move, see it not work for the teacher during competition next week, and still think it's valid! It's quite easy to watch somebody on YouTube telling you how to make $1 million, but you'll never be able to verify if they have actually achieved that themselves.
However, when you study under someone who competes, it is very easy to see if the techniques they teach you are the same ones they use in competition. If they are not, something might be up. Why? Because there are plenty of Jiu Jitsu instructors on Instagram and YouTube who showcase flashy moves that they have never used in competition, and probably don't work.
Learn from people actively doing the thing 'in competition'¶
You should always try to learn from people who are actively doing the thing that you want to do. There's some nuance too around learning from a range of people and skills too. Learning from a group of ten white belts is probably going to be a terrible and difficult path to learning, but having a mix of black belts might also be technically difficult to learn from.
If you want to succeed in business, you should learn from people who are actively engaged in business. They might not be investors, and they might not be Zuckerberg-level either. You need to find the right blend of individuals who are at your level, a few months ahead, and significantly more advanced, who are actually pursuing the things you want to do in the current moment. There are many teachers who are out of touch or too far removed from when they were students.
Rocket League is a video game where you play soccer with cars that can fly. The ability to fly is a critical aspect as it allows for a greater range of movement in the air, effectively enabling gameplay in six dimensions. It's difficult to explain, but I can share a link of a professional player demonstrating the game.
It took me two years of BJJ to appreciate the importance of strong fundamentals. When I picked up Rocket League during the pandemic, I decided to focus solely on the basics. Initially, I thought this would involve mastering the aerial flying aspect of the game. However, after watching numerous tutorials on YouTube, I discovered that the key was to drive fast and consistently hit the ball. Duh! All the tutorials unanimously advised against learning how to get airborne, as it often led to unforced errors. By the fourth week of playing, dedicating around 30 to 40 minutes a day, I had already reached the Diamond rank. I never tried to leave the air, and usually when I did, I would get scored on...
Focus on the Fundamentals, Avoid Unforced Errors¶
Half the time all of life is just avoiding unforced errors. I've realized that much of my success, my entire life, comes down to doing the simple and obvious things for a long time. I've witnessed, game after game, low-ranked players who attempted fancy moves and failed, allowing me to capitalize on their mistakes.
The road to success can be incredibly mundane. Interestingly, Rocket League was the first time I truly embraced this notion from the start. I told myself that I wanted to win, and then I started winning. I discovered that I could achieve a reasonably high level of success by simply doing two things well: steering effectively and making good contact with the ball. It's crucial not to overextend and commit unforced errors. I am convinced that most people fail to reach their desired goals due to unforced errors alone.
Here are a few examples of silly unforced errors:
- If someone wants to get jacked, they simply do not eat enough.
- If someone wants to raise money, they don't do cold outbound.
- If someone wants more followers, they only tweet once a week.
I hope some of these stories can help drive home the point about the process and thinking behind how I now pursue new hobbies.
Environments Matter: I grew up in a place with very few distractions, and it was very easy to learn. Identifying the right environment is crucial for learning. Changing your environment can be a great way to learn new things.
Get Ahead and Teach: If you're able to get ahead of your peer group, you can teach them and learn from them at the same time. If you can quickly become a valuable resource to a group of people, you can learn more just by the nature of the new and interesting questions they ask you.
Process vs Outcome: Ask yourself if the goal you're setting is an outcome or a process. If it's an outcome, ask yourself if you're measuring the right thing. If it's a process, ask yourself if you're measuring the right thing and if the order of magnitude is correct.
Periodize your training: If you're learning something that requires a lot of volume, you should consider periodizing your training. Set long time horizons and peak at the end of them. Take breaks and deload weeks.
Learn from people actively doing the thing 'in competition': Ask yourself if the people you are learning from and getting advice from are actively doing the thing you want to do. Not by proxy, or by status, or celebrity, but actually doing the thing you want to do. Then ask yourself how far along they are in their journey. Are they a few months ahead, a few years ahead, or decades ahead? Are they at risk of being out of touch?
Focus on the Fundamentals to avoid unforced Errors: Focus on the fundamentals and avoid unforced errors. Most people fail to reach their desired goals due to unforced errors alone. You should at least be failing due to circumstances you cannot control.